Learning from the Fly Photoreceptor on How Synapses Integrate Gene Expression to Sustain Retina and Brain Function

Abstract

Synapse dysfunction is thought to contribute to neurodegenerative diseases. In this issue of Neuron, Tsai et al. (2019) uncover how membrane phospholipid biosynthesis regulates Drosophila photoreceptor (PR) degeneration and the synaptic vesicle pool through a transcriptional-translational feedback loop from the synaptic terminal to the nucleus.

Retinal degenerative diseases (e.g., age-related macular degeneration, retinitis pigmentosa) and other neurodegenerative diseases like Alzheimer’s (AD) and Parkinson’s (PD) disrupt sight and cognition, respectively, affecting the central nervous system at both the cellular and synaptic levels. Experience, injury, or onset of these diseases alters synaptic function, at least in part involving presynaptic events. However, the regulation of presynaptic membrane proteins and lipids that encompass synaptic vesicles and their function in activity-dependent synaptic plasticity and neurodegeneration are poorly understood. Indeed, dysregulation and loss of synapses are early events in AD and PD cognitive decline (Hong et al., 2016). Likewise, photoreceptor (PR) synapses are impaired in retinal degeneration (Bennett et al., 2014). One example is the conditional deletion of ELOVL4, the elongase that forms very-long-chain polyunsaturated fatty acids (VLC-PUFAs, ≥C28), that results in abundance reduction of these fatty acids, synaptic abnormalities, and death of rod PRs (Bennett et al., 2014).

In this issue of Neuron, Tsai et al. (2019), using adult Drosophila PRs, demonstrate a role for presynaptic lipid metabolism in regulating genetic programs that activate a feedback loop that controls the synaptic vesicle pool and degeneration. They show that an enzyme of the biosynthesis of phospholipids, phosphatidylethanolamine cytidylyltransferase (pect), is located in the axon terminal of adult fly PRs (see Figure 1) and that it downregulates expression of sterol regulatory element binding protein (SREBP). SREBP is a member of the bHLH-zipper family of transcription factors, which sense lipid environment changes and, in turn, modulate gene expression of enzymes of fatty acid and cholesterol biosynthesis to up-hold lipid homeostasis. However, in the present study, SREBP activation, by changes in phospholipid abundance, decreases the expression of genes not directly involved in lipid metabolism, thus uncovering a novel SREBP function with loss of synaptic vesicles. These genes include four tetraspanins (TSPs). TSPs comprise a large superfamily of proteins engaged in organizing cell interactions, membrane compartmentalization, trafficking, conforming complex protein networks (“tetraspanin webs”), function of other membrane proteins (Termini and Gillette, 2017), and neurological and retinal degeneration (Gauthier et al., 2017; Termini and Gillette, 2017; Zulliger et al., 2018). Tsai et al. (2019) show that by restoring expression of two of these TSPs, SREBP activation is abolished, indicating that TSPs are functional effectors of the transcription factor in adult fly PRs. Based on these findings, they suggest a feedback loop that operates from the synaptic terminal to the nucleus linking the abundance of specific membrane phospholipids to PR function and synaptic vesicle number. They also provide evidence that TSPs are necessary for synaptic vesicle stability, endocytosis, and exocytosis and show that mutations that disrupt phospholipid biosynthesis cause loss of synaptic vesicles and that SREBP induction is necessary and sufficient to trigger synaptic vesicle loss (Figure 1). Moreover, they disclose a separate molecular mechanism for axonal degeneration that is neural-activity dependent and requires calcium for synaptic vesicle fusion.

Figure 1. Fly and Human Retinas.

The fly retina, or compound eye, has some 750 modules (ommatidia) that each contain eight photoreceptor cells (PRs). Six cells signal through synapses onto neurons of the first optic ganglion, the lamina, and the R7 and R8 PRs synapse onto neurons in the second optic ganglion, the medulla. The corneal facet and pseudocone at the top direct light down through the villi of the PR rhabdomere. Pigment cells surround each ommatidium. Capitate projections from glial cells within the lamina interact with PR terminals. A cross section through a single ommatidium shows the radial distribution of PRs, with the visual pigment-containing villi projecting inward toward the center. Six PR terminals synapse onto postsynaptic projections from the nearby lamina neurons (red circle) while two PRs pass through the lamina (cross section) to the second synaptic layer. Fly PRs form a “T synapse” onto lamina cells (illustrated within the red box). Pect → → PL (phospholipid)—l SREBP—Tetraspanins are depicted. Rod (left) and cone (right) cells within the human retina show similar cell associations. They form a “ribbon synapse” onto second-order neurons, such as horizontal cells (HCs). The ribbon synapse occurs at the tip of the human PR, indicated by the black circle. The central space within the fly ommatidium (yellow region) is analogous to the space around the tips of the human PRs (yellow).

Mutated pect disrupts phospholipid synthesis, reduces vesicle density, and activates SREBP, which triggers the loss of vesicles, capitate projections (the source of the histamine transmitter; see Figure 1), and T-bar profiles but increases endoplasmic reticulum and dense core vesicles, suggesting cellular stress. Reduced synaptic vesicle fusion limits membrane addition to plasma membrane (exocytosis) and the ability to acquire new vesicle membrane by endocytosis, inducing neuronal degeneration. Changes in phospholipid production modify membrane properties that can impair vesicle biogenesis and alter synaptic transmission, which, therefore, could protect neurons from degeneration triggered by excess demand for vesicle biogenesis.

PRPH2 (Peripherin2)/RDS (retinal degeneration slow) and ROM-1 are interacting TSPs specifically localized in the rims of the PR outer segments (Zulliger et al., 2018). They are essential for sight since more than 100 known mutations in these genes cause macular degeneration or retinitis pigmentosa (http://www.retinainternational.org/files/sci-news/rommut.htm;http://www.retina-international.org/files/sci-news/rdsmut.htm).

A dysfunctional endosomal pathway, enlarged early neuronal endosomes, and enhanced exosome secretion are characteristic of Down syndrome and AD. It is of interest that the increase of another TSP, TSP CD63, lessens endosomal abnormalities in Down syndrome, opening the possibility that controlling events of TSP availability might have potential therapeutic applications for neurodegenerations with endosomal pathology (Gauthier et al., 2017).

Like any outstanding study, the paper of Tsai et al. (2019) raises questions opening avenues of future exploration. Tsai et al. (2019) put forth the hypothesis that the endocytic defect in pect flies is due to a rate-limiting requirement for VLC-PUFA that prevents activation of the transcription factor SREBP, which, in turn, switches on protein expression, including TSPs. These fatty acids are located in synaptosomes from the inner plexiform retina layer and have been suggested to regulate shape and size of synaptic vesicles and synaptic transmission in rod PRs (Bennett et al., 2014). The formation of the VLC-PUFA derivatives, docosanoids and elovanoids, and their potential as signaling molecules should be evaluated as participants of presynaptic regulatory mechanisms, since they are known to be protective against uncompensated oxidative stress conditions in neurons and retinal pigmented epithelial cells (Bazan, 2018; see Figure 1).

To unravel the meaning of the transcriptional targets in these mutants, in terms of the functional consequences of this depletion on neurotransmission, an assessment of evoked neurotransmitter release would be of interest. One route to accomplish this involves establishing how vesicle recycling and neurotransmitter release are affected when vesicle depletion and degeneration occur due to pect mutation and when synaptic vesicle depletion due to SREBP mutation takes place. In simplest terms, it would be envisaged that there would be a decrease in evoked neurotransmitter release in these mutants. It would be of interest to define how neurotransmitter release (and indeed vesicle recycling) is affected in both the pect mutant (which displays vesicle depletion and degeneration) and the SREBP mutant (just synaptic vesicle depletion).

Revealing the molecular principles of how synaptic membrane phospholipids and proteins interact to regulate presynaptic function is a next challenge. It will entail defining the significance of phospholipids, and other lipids, beyond classical membrane “fluidity,” in membrane structural organization of synaptic vesicles and of the presynapse. It will be important to shed light on the role of presynaptic membrane lipids when synapses are selected for elimination or pruning, as during development, aging, and neurodegenerative diseases, as well as on the identification of membrane lipids that store precursors of biologically active mediators (Bazan, 2018). Understanding how they are released (and their targets), particularly lipid-mediated signaling on the neuronal transcriptome, will lead to further mechanistic insight on how neuronal survival and proper physiological function is sustained. Thus, potential targets amenable to therapeutic intervention for early onset events in brain injury (e.g., stroke and traumatic brain injury) and neurodegenerative disease affecting the nervous system will emerge. The TSP PRPH2/RDS and ROM-1 mutations that cause macular degeneration or retinitis pigmentosa could be interrogated in terms of the regulatory pathway identified in the fly. Moreover, these TSPs sustain the high curvature of disc PR edges, and it is conceivable that they might also play a similar role in the fly rhabdomeres and microvilli (see Figure 1). Thus, the testing of the fly lipid-transcription feedback loop (Tsai et al., 2019) in vertebrate retinas may expand the understanding of the regulation of these important proteins for sight and reveal mechanisms of PR outer segment biogenesis and ectosome formation (Salinas et al., 2017). Generalizing the findings from Tsai et al. (2019) to other contexts, it has been shown that dopaminergic synaptic terminal dysfunction precedes nigral cell death in PD, contributing to early clinical manifestations of disease (Bridi and Hirth, 2018; Lee et al., 2017). Whether and how lipid-transcription synaptic vesicle feedback loops exist in the context of broader neurodegenerative diseases will be of interest for future studies. Indeed, after several disappointing clinical trials targeting Aβ peptide for AD, one of the translational priorities is to develop early effective strategies to direct therapies to the roots of irreversible synaptic circuit demise. While many of these issues remain unclear, the in-depth characterization of presynaptic molecular mechanisms and the regulation of TSP expression will provide a solid basis for future studies.